Methyl-coenzyme M reductase from methanogenic archaea: Isotope effects on label exchange and ethane formation with the homologous substrate ethyl-coenzyme M

Article abstract of DOI:10.1021/ja4064876

Ethyl-coenzyme M (CH₃CH₂-S-CH₂CH ₂-SO₃^-, Et-S-CoM) serves as a homologous substrate for the enzyme methyl-coenzyme M reductase (MCR) resulting in the product ethane instead of methane. The catalytic reaction proceeds via an intermediate that already contains all six C-H bonds of the product. Because product release occurs after a second, rate-limiting step, many cycles of intermediate formation and reconversion to substrate occur before a substantial amount of ethane is released. In deuterated buffer, the intermediate becomes labeled, and C-H activation in the back reaction rapidly leads to labeled Et-S-CoM, which enables intermediate formation to be detected. Here, we present a comprehensive analysis of this pre-equilibrium. ²H- and ¹³C-labeled isotopologues of Et-S-CoM were used as the substrates, and the time course of each isotopologue was followed by NMR spectroscopy. A kinetic simulation including kinetic isotope effects allowed determination of the primary and α- and β-secondary isotope effects for intermediate formation and for the C-H/C-D bond activation in the ethane-containing intermediate. The values obtained are in accordance with those found for the native substrate Me-S-CoM (see preceding publication, Scheller, S.; Goenrich, M.; Thauer, R. K.; Jaun, B. J. Am. Chem. Soc. 2013, 135, DOI: 10.1021/ja406485z10.1021/ja406485z) and thus imply the same catalytic mechanism for both substrates. The experiment by Floss and co-workers, demonstrating a net inversion of configuration to chiral ethane with CH₃CDT-S-CoM as the substrate, is compatible with the observed rapid isotope exchange if the isotope effects measured here are taken into account.

Full text of DOI:10.1021/ja4064876

Article
pubs.acs.org/JACS
Methyl-Coenzyme M Reductase from Methanogenic Archaea:
Isotope Effects on Label Exchange and Ethane Formation with the
Homologous Substrate Ethyl-Coenzyme M
Silvan Scheller,†,§ Meike Goenrich, Rudolf K. Thauer, and Bernhard Jaun*
‡
‡
,†
†
Laboratory of Organic Chemistry, ETH Zurich, Wolfgang-Pauli-Strasse 10, 8093 Zurich, Switzerland
Max Planck Institute for Terrestrial Microbiology, Karl-Von-Frisch-Strasse 10, 35043 Marburg, Germany
‡
*
S Supporting Information
ABSTRACT: Ethyl-coenzyme M (CH CH -S-CH CH -
3
2
2
2
−
SO , Et-S-CoM) serves as a homologous substrate for the
3
enzyme methyl-coenzyme M reductase (MCR) resulting in the
product ethane instead of methane. The catalytic reaction
proceeds via an intermediate that already contains all six C−H
bonds of the product. Because product release occurs after a
second, rate-limiting step, many cycles of intermediate
formation and reconversion to substrate occur before a
substantial amount of ethane is released. In deuterated buffer, the intermediate becomes labeled, and C−H activation in the
back reaction rapidly leads to labeled Et-S-CoM, which enables intermediate formation to be detected. Here, we present a
2
13
comprehensive analysis of this pre-equilibrium. H- and C-labeled isotopologues of Et-S-CoM were used as the substrates, and
the time course of each isotopologue was followed by NMR spectroscopy. A kinetic simulation including kinetic isotope effects
allowed determination of the primary and α- and β-secondary isotope effects for intermediate formation and for the C−H/C−D
bond activation in the ethane-containing intermediate. The values obtained are in accordance with those found for the native
substrate Me-S-CoM (see preceding publication, Scheller, S.; Goenrich, M.; Thauer, R. K.; Jaun, B. J. Am. Chem. Soc. 2013, 135,
DOI: 10.1021/ja406485z) and thus imply the same catalytic mechanism for both substrates. The experiment by Floss and co-
workers, demonstrating a net inversion of configuration to chiral ethane with CH CDT-S-CoM as the substrate, is compatible
3
with the observed rapid isotope exchange if the isotope effects measured here are taken into account.
INTRODUCTION
of this reaction is unknown and lacks precedent in nonenzymic
chemistry (see the preceding publication for a more
comprehensive introduction, Scheller, S.; Goenrich, M.;
Thauer, R. K.; Jaun, B. J. Am. Chem. Soc. 2013, 135,
10.1021/ja406485z). The enzyme is highly specific for
methyl-coenzyme M. The only S-alkyl-homologue that is
known to be a substrate is ethyl-coenzyme M, which yields
■
Methyl-coenzyme M reductase (MCR) catalyzes the last step of
methanogenesis in all methanogenic archaea. The reverse
reaction is believed to be the first step in the process of
1
anaerobic oxidation of methane with sulfate as the terminal
2
,3
electron acceptor. MCR contains two molecules of the nickel
cofactor F430 as prosthetic groups, which must be in their
Ni(I) oxidation state for the enzyme to be active (Figure 1).
4,5
ethane as the product (Scheme 1).
−
MCR converts methyl-coenzyme M (CH -S-CH CH -SO ,
Me-S-CoM) and the thiol coenzyme (CoB-SH) to methane
and the corresponding disulfide, respectively. The mechanism
3
2
2
3
Intermediates in the Catalytic Cycle. The catalytic
reaction of MCR involves at least one intermediate for both the
6
native substrate methyl-coenzyme M and ethyl-coenzyme M.
The intermediate formed from ethyl-coenzyme M reacts about
1
00 times faster back to the substrate, than forward to free
ethane. From label exchange experiments in deuterated
medium it could be concluded that all six C−H bonds of
ethane must already be present in the intermediate and that
ethane is released only after a subsequent rate-limiting step.
13
The isotope exchange pattern with a mono- C-labeled ethyl
group in the substrate is depicted in Scheme 2.
Received: June 26, 2013
Published: September 4, 2013
Figure 1. Coenzyme F430, the prosthetic group of MCR.
©
2013 American Chemical Society
14985
dx.doi.org/10.1021/ja4064876 | J. Am. Chem. Soc. 2013, 135, 14985−14995

Journal of the American Chemical Society
Article
Scheme 1. Catalytic Reaction of MCR with the Homologous
Substrate Ethyl-Coenzyme M Yielding the Product Ethane
Instead of Methane
Scheme 3. Isotope exchange sequence for 13C-labeled ethyl-
coenzyme M (isotopologues in Scheme 2 are: 1, 2, 3, 4, 5, 6,
and 6′)
a
Scheme 2. Reaction of CH313CH -S-CoM with CoB-SD and
2
a
Enzyme in D O Buffer, Neglecting Ethane Formation
2
a
The intermediates are represented by crossings (or vertices) of the
connecting lines; “CoM” parts are omitted for clarity. Each reaction in
deuterated medium generates the intermediates on the next lower
level; reaction in H O buffer generates intermediates on the next
2
higher level. Shown is the numbering of isotopologues used in the
Results section of this study. Isotopomers are labeled in the same
color; 13C-isotopomers are located at vertically symmetric positions.
a
13
The formation of the intermediates (depicted in brackets) is
For experiments without C-labeling, the identical nomenclature is
dependent on the solvent. In D O buffer, a deuterium atom is
used (species 1−12, without an slanted prime).
2
introduced into the intermediate (blue arrows), whereas in H O buffer
2
a hydrogen atom would be introduced. The re-formation of substrates
the possibility to deduce β-secondary isotope effects. Our
values for the KIEs were then used in a kinetic simulation of the
Ahn and Floss experiment to judge the compatibility of their
results with our measurements.
(
black arrows) is independent of the H/D content of the solvent but is
affected by primary and secondary isotope effects involving the C−H/
C−D bonds in the intermediate.
1
3
RESULTS
With a mono- C-labeled ethyl group, 24 different
isotopologues are possible (without counting enantiomers),
which leads to the exchange sequence shown in Scheme 3.
Experiment to Deduce the Stereochemical Course of
the Reaction to Chiral Ethane. Ahn and Floss incubated the
■
1
3
Label Exchange in CH CH -S-CoM as the substrate.
Isotopologue 1, (CH3 CH -S-CoM, 10 mM) was incubated
with coenzyme B (2 mM) and active enzyme (0.94 μM) in
D O buffer at 60 °C for 2, 4, 8, 16, and 32 min. The assay
solutions were analyzed via C{ H}{ H} NMR spectroscopy
in order to quantify all isotopologues present, including those
containing a fully deuterated ethyl group (Figure 2). A time
course for each isotopologue is plotted in Figure 3.
Label Exchange in Deuterated Substrates. Deuterated
ethyl-coenzyme M was incubated in H O buffer to study the
3
2
13
2
2
1
3
1
2
chiral substrate CH CDT-S-CoM with a cell-free extract from
3
Methanosarcina barkeri in order to obtain chiral ethane
CH CHDT), which was converted to acetate and analyzed
3
7
(
for its configuration. A net inversion of configuration was
found (41% ee for the S and 51% ee for the R substrate). The
finding of inversion of configuration is mechanistically
significant, since it rules out a rotatable free ethyl radical or
ethyl cation as an intermediate. At first sight, our discovery of
rapid carbon scrambling and H/D exchange in the substrate
ethyl-coenzyme M appears to be incompatible with the results
obtained by Ahn and Floss. However, the competition between
wash-out of deuterium (or tritium) and the formation of ethane
depends on both conversion and potential isotope effects.
Here, we present the results of our study of all kinetic isotope
effects (KIEs) involved in C−H bond formation and activation
for the substrate ethyl-coenzyme M, which were extracted from
kinetic simulations fitted to the observed time course for the
label exchange in ethyl-coenzyme M. The determined KIEs will
be compared with those obtained with the native substrate
methyl-coenzyme M. In addition, ethyl-coenzyme M offered
2
wash out of deuterium. The isotopologues CD CD -S-CoM
3
2
(12) and CH CD -S-CoM (4) were used and two different
3
2
enzyme concentrations investigated. This set of isotope
exchange experiments enabled extraction of the isotope effects
for intermediate formation and for ethane activation in the
intermediate. All relevant experiments are summarized in Table
1.
Experiments with the Isotopologue CH CD -S-CoM As
the Substrate. The experiments with the isotopologue
3
2
CH CD -S-CoM (4) as the substrate in H O buffer (experi-
3
2
2
ments no 4 and 5) are discussed in more detail here, since
incubation with MCR in H O buffer yields a mixture of only 5
2
different species (1−5, see Scheme 4) that can all be quantified
1
accurately by H NMR spectroscopy (Figure 4).
1
4986
dx.doi.org/10.1021/ja4064876 | J. Am. Chem. Soc. 2013, 135, 14985−14995

Journal of the American Chemical Society
Article
Figure 2. Expansion of the 150 MHz 13C{ H}{ H} NMR spectra (simultaneous proton and deuterium broad-band decoupling) of the assay
1
2
solutions before (“0 min”) and after incubation with active enzyme at 60 °C for the time indicated. The numbering of isotopologues corresponds to
13
that in Scheme 3. Lines marked with * are due to natural abundance of C at the “unlabeled” position within the ethyl group. The signal marked
13
13
−
with ** is due to natural abundance of C at carbon 2 (Et-S- CH CH -SO ). The fraction of isotopologues 4′ and 7′ remained below the
2
2
3
detection limit.
13
Figure 3. Time course of all 24 isotopologues during incubation of CH CH -S-CoM with MCR in deuterated medium. The molar fractions of the
3
2
isotopologues were obtained via integration of the 13C{ H}{ H} NMR spectra depicted in Figure 2 (inverse gated, 60 s relaxation delay). The
isotopologues were normalized to 100% for each time point. The conversion to ethane at the last time point (1920 s) was about 5% as deduced from
the amount of disulfides formed.
1
2
1
2
The H{ H} NMR spectrum (Figure 4) proves that only the
expected isotopologues are present. The time course for each
isotopologue (1−5) is plotted in Figure 5.
in the C−H/C−D-activation in the intermediate [CH CHD ]:
3
2
5
is formed ∼3 times more rapidly than 2, while a ratio of 3:2
would be expected from statistics alone (number of H- and D-
atoms, respectively).
The isotope exchange pattern described in Scheme 4
matches the experiment. The initial substrate (isotopologue
) is first converted to isotopologues 5 and 2. In a second step,
isotopologues 3 and 1 appear. Visual inspection of Figure 5
already shows that substantial isotope effects must be involved
Kinetic Modeling. In order to obtain the isotope effects
involved in this exchange reaction, the sequence was simulated
4
8
with the program COPASI, (see Supporting Information [SI]
1
4987
dx.doi.org/10.1021/ja4064876 | J. Am. Chem. Soc. 2013, 135, 14985−14995

Journal of the American Chemical Society
Article
Table 1. Experiments Performed to Obtain a Time Course
for Each Isotopologue
a
b
exp. no
substrate
solvent
enzyme [μM]
3
1
1
2
3
4
5
6
7
a
CH31 CH -S-CoM (1)
D O
0.94
0.94
0.94
3.75
0.94
3.75
0.94
3.75
2
2
c
CH313CH -S-CoM (1)
b
c
D O
2
2
CH CH -S-CoM (1)
D O
2
3
2
CH CH -S-CoM (1)
D O
2
3
2
CH CD -S-CoM (4)
H O
2
3
2
CH CD -S-CoM (4)
H O
2
3
2
CD CD -S-CoM (12)
H O
2
3
2
CD CD -S-CoM (12)
H O
2
3
2
a
Each experiment consisted of six samples that were incubated for 0, 2,
4
, 8, 16, and 32 min, respectively at 60 °C. Exp. 1a was analyzed via
1
3
1
2
1
2
b
C{ H}{ H} NMR, exp. 1b−7 were analyzed via H{ H} NMR. c =
c
6
10 mM. Data reported earlier and included in the kinetic simulation
of isotope exchange reported here.
Scheme 4. Isotope Exchange Pattern for CH CD -S-CoM (4,
3
2
a
box) in H O Buffer
1
2
2
Figure 4. Expansion of the 600 MHz H{ H} NMR spectrum showing
the methyl-group of Et-S-CoM after incubation with MCR for 32 min
(exp. no 4). (Black line) Acquired spectrum. (Red line) Simulated
spectrum. Assignments follow the numbering in Schemes 3 and 4
a
Intermediates are represented in brackets. (Red pathways) Specific to
H O buffer; the blue pathways would take place in D O buffer
2
2
(
dotted). Re-formation of substrates (black pathways) depends solely
on the isotope effects involved.
S3.1 for a comprehensive description). The catalytic cycle was
modeled in four reaction steps (Scheme 5).
For each step (Scheme 5), a rate constant was defined for
both directions. Binding of the second substrate coenzyme B
Figure 5. Time course of all isotopologues (1−5) obtained after
incubation of CH CD -S-CoM in H O buffer containing coenzyme B
and active enzyme.
3
2
2
(
step 2) defines whether hydrogen or deuterium will be
introduced in the next step and ultimately depends on the
9
isotopic composition of the buffer used (H O or D O).
The rate of release of ethane (step 4, kprod) was adjusted to
match the experiment and assumed to be independent of the
isotopic constitution (see below for isotope effects on ethane
release). To compare the experimental data more accurately
with the simulation, all isotopologues that could be quantified
2
2
For the catalytic transformation (step 3, k ), deuterium atoms
f
in the CH - and -CH - positions of the ethyl group give rise to
3
2
α and β secondary isotope effects on intermediate formation,
which were both taken into account in the simulation. For the
1
3
12
13
13
1
2
substrates labeled with C, a C/ C KIE was included in the
simulation.
(all for C{ H}{ H} NMR analysis and only the H-containing
1
2
ones for H{ H} NMR analysis) were normalized to 100% for
both the experiment and for the simulation (see Experimental
Procedures). The primary isotope effect (whether hydrogen or
deuterium is introduced) corresponds to the ratio of enzyme
activities in H O vs D O and was set to 1.0 (see below).
In the reverse reaction (activation of a C−H/C−D bond of
the ethane moiety in the intermediate; step 3, k ), the same set
b
of isotope effects was considered. In addition, the number of
equivalent H or D leading to the same reformed substrate had
to be taken into account as depicted in Scheme 6 for the first
2
2
Because of a gradual enzyme deactivation with increasing
turnovers (later time points) and a delay in warming up to 60
intermediate formed from isotopologue 1 in D O buffer.
2
1
4988
dx.doi.org/10.1021/ja4064876 | J. Am. Chem. Soc. 2013, 135, 14985−14995

Journal of the American Chemical Society
Article
a
Scheme 5. Reaction Steps Simulated for the Isotope Exchange (steps 1−3) and for Ethane Formation (step 4)
a
Step 1: binding of the first substrate ethyl-coenzyme M. Step 2: binding of the second substrate coenzyme B. Step 3: catalytic step transforming the
substrates ternary complex to the intermediate containing ethane (k , k ). Step 4: This step is rate limiting and was simulated as a single elementary
f
b
step with simultaneous release of both products (kprod).
Scheme 6. Isotope Effects for C−H/C−D Bond Activation in
the Intermediate That Were Taken into Account in the
Kinetic Simulation, Exemplified for the Intermediate
According to the kinetic model, all isotope effects have to be
identical for all experiments. The values determined independ-
ently were averaged to give the values listed in Table 3, and a
conservative confidence interval was estimated.
1
3
Formed from CH CH -S-CoM (1) as the Substrate in D O
3
2
2
a
buffer
Table 3. Isotope Effects Averaged from All Simulations,
a
Including the Estimated Confidence Intervals
isotope
effect
intermediate
formation
C−H/C−D activation from
intermediate
b
primary
1.0
2.4 ± 0.2
α (per D)
β (per D)
1.22 ± 0.10
1.0 ± 0.05
1.17 ± 0.10
1.05 ± 0.05
a
b
See Reliability of Isotope Effects below. See below for
a
The rate constant for INT formation (k ) and the rate constant for
f
determination.
substrate re-formation (k ) are modified through all isotope effects
involved (specified in parentheses after the constant). 1° and
b
13
C
Now, each experiment (1−7) was resimulated with the
unified set of isotope effects (listed in Table 3) obtained
through averaging over all experiments. A graph comparing
each experiment with the unified simulation is provided in S3.3
in SI.
signify primary KIEs, α and β signify secondary KIEs. The rate of C−
H/C−D bond activation (k ) is multiplied by the number of
b
equivalent H/D atoms present.
Reliability of the Isotope Effects Determined via
Simulation. The value of each isotope effect was lowered and
increased by the confidence interval estimated (Table 3) in
order to determine its sensitivity to this parameter. Experiment
°
C (2-min time point), the experimental time scale had to be
adjusted to obtain a constant reaction rate, enabling
comparison with the simulation. Enzyme deactivation with
1
increasing turnovers is commonly observed with MCR. The
relative time adjustment for each time point for all experiments
is provided as S4.1 in the SI.
5
(reaction of CH CD -S-CoM (4) in H O at high enzyme
3 2 2
concentration) was utilized for this purpose, because a double
secondary isotope effect is involved (square of the intrinsic α-
KIE). In addition, only five different species are formed in this
experiment (see Scheme 4) that could all be determined
Extraction of Isotope Effects via Kinetic Modeling of
the Exchange Reactions. Without isotope effects, the
simulation could not be matched to the experiment, which
signifies that IEs must be present and that step 4 is indeed rate-
limiting. See S3.2 (in SI) for the simulation of experiment 5
from Table 1 without isotope effects (Figure S3.3 in SI) and for
further discussion.
For each of the seven experiments, the combination of
isotope effects yielding the best overall fit to the experiment was
evaluated independently. For experiment 1, the analysis was
1
accurately via H NMR spectroscopy.
Variations of the α-secondary KIE for intermediate formation
and the primary KIE for C−H/C−D bond activation in the
intermediate are depicted in Figure 6. Variations of all other H/
1
3
D KIEs are provided in S3.4 in SI, the variations of C-KIEs
are tested in experiment 1b (see S3.5 in SI for simulations with
and without 13C KIE).
1
3
1
2
1
2
The variations by the confidence intervals clearly result in a
much poorer fit, which shows that the (rather large) confidence
intervals estimated are reliable. Note that a large variation in the
molar fraction of a specific isotopologue upon changing a
particular isotope effect means that the simulation is sensitive to
that isotope effect. Accordingly, the band between the thin line
and the dotted line in Figure 6 can be regarded as the inverse of
an error band.
carried out using C{ H}{ H}- and H{ H} NMR data
independently. The best parameters extracted for each
experiment are listed in Table 2.
Table 2. Individual Sets of Kinetic Isotope Effects Matching
Each Experiment As Closely As Possible
a
a
KIE
1a
1b
2
3
4
5
6
7
Primary Isotope Effect on Intermediate Formation.
The primary isotope effect on intermediate formation cannot
be measured in a competitive experiment since it corresponds
to the rate depression observed when the reaction is carried out
in D O (experiments 1−3) instead of H O (experiments 4−7).
α(kf)
β(kf)
1.2
1.0
2.4
1.2
1.0
1.25
1.0
1.3
1.0
2.4
1.15
1.0
1.2
1.0
2.4
1.15
1.0
1.2
1.2
1.0
2.4
1.2
1.1
1.2
1.0
2.4
1.15
1.0
1.2
1.0
2.4
1.15
1.0
1.0
1°(kb)
2.5
2.4
α(kb)
β(kb)
1.15
1.0
1.15
1.05
2
2
However, it could be estimated from the ratio of the average
simulated times vs experimental times, which correspond to the
relative enzyme activities.
a
A 13C KIE of 1.05 was assumed for k and for k , since a better fit to
the experiment was obtained (see S3.5 in SI)
f
b
1
4989
dx.doi.org/10.1021/ja4064876 | J. Am. Chem. Soc. 2013, 135, 14985−14995

Journal of the American Chemical Society
Article
Figure 6. Overlay of measurements (points) and simulations (lines) for experiment 5, including the lower and upper limits for the isotope effects
determined. (a) Variation of the α-secondary isotope effect on intermediate formation by the estimated confidence interval. (Solid line) Best value
determined: KIE = 1.22. (Thin line) KIE = 1.32. (Dotted line) KIE = 1.12. (b) Variation of the primary isotope effect for the activation of the C−H/
D bond in the ethane-intermediate by the estimated confidence interval. (Solid line) Best value determined: KIE = 2.4. (Thin line) KIE = 2.6.
(
Dotted line) KIE = 2.2). Corresponding variations of all other KIEs can be found in S3.4 (in SI). Note that the experimental times were adjusted in
order to account for activity during warm-up and the enzyme deactivation generally observed with MCR (see S4.1 in SI for details).
The average activity for the assays in H O was found to be
moiety and that a possible steric isotope effect on the
2
equal to that in the assays in D O, but the dispersion is large
populations of the rotamers of the -CH D group is too small
2
2
due to enzyme deactivation via oxidation of the Ni center in the
enzyme (see S4.1 and S4.2 in SI). Because of the experimental
scatter associated with enzyme activity, the primary kinetic
isotope effect can only be delimited to the range 1.0 ± 0.3.
Testing for Enantioselectivity in the Formation of
to be detectable (see S6.3 in SI for a detailed analysis).
Isotope Effect on the Formation of Free Ethane from
the Intermediate. A potential isotope effect on step 4 was
investigated by measuring the relative rate of ethane formation
from the CH CH intermediate and from the CD CHD
2
3
3
3
CH CHD-S-CoM. Substrate isotopologues with one D and one
3
intermediate. To generate such a mixture of intermediates, a
mixture of CH CH -S-CoM and CD CD -S-CoM was
H in the methylene group are chiral. We investigated whether
3
2
3
2
substrate re-formation to CH CHD-S-CoM (2) shows any
3
dissolved in H O buffer and incubated with CoB-SH and
enantioselectivity. Unlabeled ethyl-coenzyme M was incubated
2
active enzyme. The IEs for step 3 influence the actual ratio of
the two intermediates and have to be taken into account in
order to calculate the IE for step 4 (Scheme 7).
in D O buffer with CoB-SH and MCR to yield ∼20%
2
conversion to monodeuterated isotopologues (CH DCH -S-
2
2
CoM (3) and CH CHD-S-CoM (2)). After oxidation to
3
sulfoxides, the ratio of the enantiomers for CH CHD-S-CoM
A high concentration of ethyl-coenzyme M (200 mM) was
used in this experiment in order to measure the initial rate and
minimize relative isotope exchange. The ratio of undeuterated
to pentadeuterated ethane found in the head space was
3
2
was analyzed by H NMR using (−)-dihydroquinine hemi-
sulfate for chiral resolution (for details and spectra see S6 in SI)
Additionally, experiments with CD CD -S-CoM in H O and
3
2
2
with CD CH -S-CoM in D O, which both lead to the chiral
1
2
3
2
2
analyzed via H{ H} NMR spectroscopy. The small amount of
CHD CHD and CD CH D formed was added to the amount
of CD CHD in the calculation of the isotope effect (Table 4).
isotopologue CD CHD-S-CoM (10), were carried out and
3
1
2
2
2
3
2
investigated in the same way but via H{ H} NMR spectros-
copy (data not shown). In all three experiments, the chiral
isotopologues were formed as racemic mixtures. In the
intermediate, the two hydrogens of the enzyme-bound
CH CH D group are diastereotopic, in principle. The fact
3
2
In the absence of an IE for step 4 (kprod), a ratio CH CH /
3
3
CD CHD of 1.49/2.89 = 0.515 would be expected for the
3
2
overall reaction from an equimolar mixture of the two
substrates to ethane due to the KIEs on step 3 (see S4.3 in
SI for a detailed description).
3
2
that C−H activation occurs at the same rate for H and HSi
Re
confirms that the intermediate contains a complete ethane
1
4990
dx.doi.org/10.1021/ja4064876 | J. Am. Chem. Soc. 2013, 135, 14985−14995

Journal of the American Chemical Society
Article
Scheme 7. Reaction Scheme for the Experiment To
Scheme 8. Kinetic System for the Simulation of the Loss of
% ee Due to Isotope Exchange
a
Determine the Isotope Effect on Ethane Formation with All
a
Isotope Effects Involved
a
The ratio CH CH /CD CHD (gas in the headspace) was measured.
3
3
3
2
Red: the isotope effect on ethane formation from the intermediate,
kprod(d )/k (d ), can be calculated, because all isotope effects for k
f
0
prod
5
and k are known from the simulation.
b
a
Shown is the case starting with the (R)-substrate (box).
The isotope effect on kprod (d vs d ) corresponds to the
measured ratio (0.63) divided by the ratio expected without an
IE on kprod (0.515):
0
5
In the simulation, the commitment (kprod/4k in Scheme 8)
b
was set to 0.00926, corresponding to our results obtained with
the pure enzyme. The basic rate constant was set to give a total
conversion of ∼25% within 10 h (in Floss’ experiments
incubation with cell-free extract overnight gave conversions of
0.63 ± 0.03/0.515 ± 1.22 ± 0.06.
Taking the errors for the KIEs on k and k into account, the
f
b
2
8% for the R-substrate and 10.3% for the S-substrate).
errors accumulate, and we arrive at an IE for kprod (d vs d ) =
0
5
Additional simulations were carried out, assuming only partial
stereospecificity for the formation of the intermediate and re-
formation of the substrate (k and k ). The result of the
1
.2 ± 0.2.
Simulation of the Experiment by Floss and Co-
f
b
workers with the KIEs Determined in This Study. Using
the KIEs obtained in this study we subsequently simulated the
experiments determining the stereochemical course for ethane
simulations is shown in Figure 7.
7
formation reported by Ahn and Floss. A “chemical” kinetic
model without the substrate-binding steps was simulated with
the COPASI program (Scheme 8).
Both the forward and reverse-to-substrate reactions (k and kb
f
in Scheme 8) were assumed to occur under inversion of
configuration. Initially, there are four species (2 mM total
according to ref 7): (R)-CH CDT-S-CoM, (S)-CH CDT-S-
3
3
CoM, (R)-CH CHD-S-CoM, and (S)-CH CHD-S-CoM. Their
3
3
concentrations were set according to the % ee estimated for the
substrate in ref 7 (75% max) and to the activities (1.6 and 2.8
μCi) reported. Most of the substrate is (R)-CH CHDSCoM
3
and (S)-CH CHD-S-CoM (where H replaces T), again in a
3
ratio corresponding to the % ee of 75%. The product
CH CH T is counted but is achiral. In the F-value analysis of
3
2
acetate, this species would be indistinguishable from racemic
[
1,1-D,T]-ethane (F value near 52%).
The % ee of formed ethane therefore corresponds to:
Figure 7. Simulation of the loss of % ee due to isotope exchange,
taking into account the KIEs reported here and assuming complete
inversion of configuration (solid black curve), and experimental values
%
ee ± {|[(R)‐CH CHDT] − [(S)‐CH CHDT]|}
3
3
of ref 7 (red dots) for the two enantiomeric substrates (R)-CH CDT-
3
/
{[(R)‐CH CHDT] + [(S)‐CH CHDT]
3
3
S-CoM and (S)-CH CDT-S-CoM. Colored curves: simulations
3
assuming only partial stereospecificity as indicated by inversion %.
+
[CH CH T]}
3
2
Table 4. Isotope Effect on Ethane Formation Starting with a Mixture of Unlabeled Ethyl-Coenzyme M (1) and CD CD -S-CoM
3
2
a
(
12) As the Substrates
time [min]
CH CH [mol %]
CD CH D [mol %]
CHD CHD [mol %]
CD CHD [mol %]
sum d /d [mol %]
ratio d /(d + d )
0 5 4
3
3
3
2
2
2
1.1
1.6
2.5
3
2
4
5
2
4
8
38.1
38.8
39.0
1.1
59.7
57.9
55.9
61.9
61.2
61.0
0.615
0.635
0.641
1.6
2.5
a
The ratio unlabeled/deuterated ethane measured at the three different incubation times averages to 0.63 ± 0.03.
1
4991
dx.doi.org/10.1021/ja4064876 | J. Am. Chem. Soc. 2013, 135, 14985−14995

Journal of the American Chemical Society
Article
Whereas washout of deuterium leads to loss of chirality
or absent KIE on TS2 does not allow us to distinguish between
the two possible structures discussed here for the intermediate.
Radical Mechanisms. Since the IEs determined for ethyl-
coenzyme M are very close to those found for methyl-
coenzyme M, our assessment that the experimental IEs are
qualitatively compatible with the radical-substitution mecha-
nism of Siegbahn and Crabtree (Scheller, S.; Goenrich, M.;
Thauer, R. K.; Jaun, B. J. Am. Chem. Soc. 2013, 135, 10.1021/
ja406485z) also holds for ethyl-coenzyme M. As with Me-S-
CoM, a more quantitative comparison will require a DFT
calculation for Et-S-CoM on the same level of theory and
including vibrational analysis. It would be of particular interest
whether such a calculation predicts equal activation energies for
step 3 (TS1) for the two substrates as found by experiment
(
CH CDT-S-CoM is replaced by CH CHT-S-CoM, which
3
3
gives CH CH T as product), any partial nonstereospecificity in
3
2
the formation of the intermediate and re-formation of substrate
would lead to fast racemization of the substrate because, for
each molecule going to product, 100 molecules are reacting
back to substrate. In the simulations, this can be seen as a large
initial drop of the % ee (colored curves in Figure 7).
Probing Propyl- and Allyl-S-CoM for Deuterium
Incorporation. Propyl-coenzyme M and allyl-coenzyme M
are not converted to propane or propene, respectively, but are
known to be competitive inhibitors of MCR binding in the
5
active site. We investigated both inhibitors for deuterium
incorporation analogously to the experiments with ethyl-
coenzyme M, but no deuterium incorporation could be
observed for either inhibitor, even at high enzyme concen-
trations.
(
see below). No explanation for the observed difference in
energy of TS2 relative to TS1 for the two substrates is derivable
from the reported DFT calculation because the alkane was
removed from the structural assembly for the calculation of the
DISCUSSION
17
■
second step. Hence, radical back-substitution of CoB-S• on
Isotope Effects for Label Exchange in Ethyl-Coen-
zyme M. The KIEs for isotope exchange in ethyl-coenzyme M
reported here correspond closely to those obtained for methane
formation and methane activation, a strong indication that the
reaction mechanism is the same for the two substrates.
However, as reported earlier, different steps are rate-limiting
if the substrate is methyl-coenzyme M (see preceding article) or
the sulfur of Ni-coordinated coenzyme M thiolate and
formation of the heterodisulfide and Ni(I)F430 is identical
for both substrates.
The fact that no deuterium is incorporated into the
competitive inhibitors allyl-coenzyme M and propyl-coenzyme
is puzzling because their binding induces the same subtle
changes in the EPR spectrum that are observed when Me-S-
6
ethyl-coenzyme M. In order to compare the KIEs in the first
CoM is added to active MCR, thus confirming that these5
catalytic step for the two substrates, one has to study methane
formation in the case of methyl-coenzyme M but isotope
exchange into the substrate for ethyl-coenzyme M.
inhibitors bind in the active site like the natural substrate.
app
Since K (allyl-coenzyme M) = 0.2 mM is lower than
i
app
5
KM (Me-S-CoM) = 5 mM, the active site of free MCR can
The primary H/D KIE for the formation of the intermediate
from substrate is small (1.0−1.3), while in the reverse direction,
the primary H/D KIE of C−H activation from the intermediate
is much larger (2.4 for both substrates). Substantial α-
secondary KIEs of ca. 1.2/D were found in both directions
for both substrates, indicating a change in hybridization from
accommodate the larger C substrates in the binary substrate
3
complex.
According to the calculated reaction profile of Chen et al.,
the late transition state of the first, endothermal step (TS1)
17
contains an almost free alkyl radical. Therefore, the energy of
TS1 should directly reflect the stability of this radical (e.g., the
BDE of the C−S bond) and the activation energy should
decrease in the order methyl- > ethyl- ≫ allyl-coenzyme M. In
order to try to explain the unreactivity of allyl-coenzyme M
within the framework of the mechanism proposed by Siegbahn
and co-workers one must assume that the larger S-alkyl
substituents sterically hinder the binding of coenzyme B or the
approach of its thiol group to the nascent alkyl radical.
3
2
sp toward sp upon going to the transition state of step 3 in
either direction.
The β-secondary KIEs (1.0 ± 0.05, 1.05 ± 0.05) found for
ethyl-coenzyme M in the two directions of step 3 are not
consistent with strong hyperconjugation in the transition state
such as would be expected for an ethyl-cation-like TS, but
congruent with the small β-secondary KIEs for homolytic
1
0
cleavage to ethyl radical found in the literature (all <5%).
1
2
13
Why Are the Steps after the Intermediate Rate
Limiting for Ethyl-Coenzyme M but Not for Methyl-
Coenzyme M? From the initial rate of methane formation
In contrast to methyl-coenzyme M, the C/ C-KIE could
not be measured directly for ethyl-coenzyme M, but the
simulations gave the best fit with k /k = 1.05, close to the
1
2
13
13
from methyl-coenzyme M and the initial rate of C scrambling
value found for methyl-coenzyme M and consistent with
breaking the carbon−sulfur bond in the first catalytic step.
Nature of the Ethane-Containing Intermediate. The
fact that all six C−H bonds of ethane are already present and
equivalent in the intermediate leaves only two structural
possibilities: either noncoordinated ethane held in the active
site by noncovalent forces or a (CH CH )Ni σ-complex. In the
in ethyl-coenzyme M the relative rates of the first catalytic step
(
step 3, TS1, k ) for the two substrates can be estimated.
f
Taking into account the different apparent K values (Me-
M
5
CoM = 5 mM; Et-CoM = 20 mM) the specific activities for
formation of the intermediate from the substrate ternary
complex are practically the same for the two substrates (Me-S-
3
3
−
1
−1
CoM: 20 U mg and Et-S-CoM: 22 U mg ; both at
latter case, the dissociation energy of the σ-complex might
contribute to the activation energy of the following, rate-
limiting step (TS2). However, from the experimental IE on the
overall reaction as determined in a competitive experiment with
CH CH -S-CoM and CD CD -S-CoM we derive a KIE on the
18
saturation). According to the kinetics of isotope exchange, the
highest point in the reaction profile leading from the
intermediate to products is 8.3 kJ mol− lower than the
transition state of step 3 (TS1) for Me-S-CoM but 12.3 kJ
mol higher than TS1 for Et-S-CoM (see Scheme 4 in ref 6).
Since both, C−S bond breaking and C−H bond formation
occur in step 3 (TS1), why are the steps after the intermediate
slower for ethyl-coenzyme M than for methyl-coenzyme M?
1
3
2
3
2
−1
rate-limiting step that is close to unity (k /k = 1.2 ± 0.2). In
H
D
11−16
contrast to equilibrium IEs,
no experimental data for kinetic
isotope effects on the dissociation of alkane σ-complexes have
been reported. In view of this lack of reference data, the small
1
4992
dx.doi.org/10.1021/ja4064876 | J. Am. Chem. Soc. 2013, 135, 14985−14995

Journal of the American Chemical Society
Article
If the intermediate contains the free alkane, these subsequent
steps comprise the formation of the heterodisulfide, which is
chemically identical for both substrates, and the release of
products. It is still unknown which one of the two products is
released first. X-ray structures of inactive Ni(II) forms of the
enzyme show a long narrow channel for entrance of the
substrates that is completely plugged by bound coenzyme B
substantiating or excluding such a complex as the intermediate
could be obtained in our study.
The kinetic simulation of Floss’ experiment showed that7,
when all isotope effects are taken into account, the reported
reduction of enantiomeric excess is fully attributable to isotope
exchange in the substrate. This indicates that each elementary
step in ethane formation is stereospecific with inversion of
configuration as the overall result.
19
(MCRox1‑silent) or by the heterodisulfide product (MCRsilent).
This suggests that methane or ethane can only leave the active
site after the heterodisulfide has been released. Rough estimates
Our finding that ethyl- and methyl-coenzyme M show the
same isotope effects, and therefore react according to the same
catalytic mechanism, enhances the significance of all experi-
ments done with ethyl-coenzyme M, among them the
for the apparent K values for methane (>10 mM) and for the
M
18
heterodisulfide (≤1 mM) are in accord with this suggestion,
7
because the K value of the first substrate is usually higher than
experiment by Floss and co-workers, which cannot be carried
M
that of the second substrate in ordered-bi bi mechanisms. In
this picture of events, the reason why a step following
intermediate formation becomes rate limiting with ethyl-
coenzyme M as the substrate would have to be steric hindrance
of either the formation or the release of the heterodisulfide by
the slightly larger ethane molecule in a very tight active site.
Compatibility of Rapid Isotope Exchange with the
Experiments Reported by Ahn and Floss. The substantial
out with the natural substrate methyl-coenzyme M.
EXPERIMENTAL PROCEDURES
■
Assay Conditions. The standard assays contained 10 mM of the
desired isotopologue of ethyl-coenzyme M, 2 mM coenzyme B and an
aliquot of MCR stock solution giving the amounts indicated.
Coenzyme B is needed only catalytically for the isotope exchange.
For the formation of ethane, the maximal fraction of conversion is
20%.
primary KIE of 2.4 for C−H bond cleavage (k ) and the low
b
conversion to free ethane reduce the effect of deuterium
washout to the point where the predicted % ee is still 57 (expt
with (S)-substrate: 41) at 10.3% and 43 (expt with (R)-
substrate: 51) at 28% conversion. Taking into account the
considerable experimental difficulties of the F-value determi-
For each set of experiments, a stock solution (50 mM potassium
phosphate buffer, pH = 7.6 in H O or D O, including the substrates)
2 2
was prepared under ice cooling without enzyme. One aliquot was
removed as a reference (“0 min value” or control). The remaining
amount was supplemented with enzyme and distributed to different
vials (1.60 mL into each). The volume of the liquid phase for all
experiments was 1.60 mL. After being sealed with a rubber stopper, the
vials were immediately incubated at 60 °C in a water bath shaker. The
reaction was stopped after the desired reaction time by addition of 1
mL air with a syringe (immediate deactivation of MCR).
20
nation reported by Ahn and the fact that these authors
studied cell free extracts containing MCR from Methanosarcina
barkeri at 37 °C whereas our KIEs were obtained with MCR
from Methanothermobacter marburgensis at 60 °C, the fast
isotope exchange in ethyl-coenzyme M and the observation of
net inversion of configuration by Ahn and Floss are not
contradictory. However, it remains unexplained why the
experimental %ee was higher for the run with higher conversion
in ref 7.
The remaining ethyl-coenzyme M was analyzed via NMR
spectroscopy for its isotopic constitution. The fraction of conversion
was determined by measuring the amount of disulfides formed. For the
experiments measuring the product ethane, the gas was passed through
1
2
CDCl3 to absorb the ethane and measured via H{ H} NMR
spectroscopy.
Isotope Exchange Experiments with Different Ethyl-Coen-
zyme M Isotopologues As Substrates. All seven experiments
described in Table 1 were performed using the standard assay
conditions described above with the same enzyme preparation. For
experiments 1, 2, 4, and 6 the amount of enzyme was 1.5 nmol per
assay (0.94 μM); for experiments 3, 5, and 7 an amount of 6.0 nmol
(3.75 μM) was used (see Table 1).
CONCLUSION
■
Within experimental error, the kinetic isotope effects
determined for C−H bond formation and C−H bond
activation are identical for the native substrate Me-S-CoM
and for its homologue Et-S-CoM, thus indicating that they react
according to the same reaction mechanism. The large α-
secondary KIEs and the fact that methyl- and ethyl-coenzyme
M react at the same rate in the C−S bond breaking step speak
For experiment 1, the analysis was carried out via integration of the
1
1
3
1
2
C{ H}{ H} NMR spectra (experiment 1a) and through fitting of the
2
H{ H} NMR spectra (experiment 1b) independently. For experiment
1
9
b, the following isotopologues were quantified: 1−6, 8, 9, 11, 6′, 8′,
against the earlier proposal of an S 2 attack on carbon giving
alkyl-Ni(III) species.
N
5
,19,21
′, and 11′. For experiments 2, 3, 6, and 7: isotopologues 1−11 (all,
The value determined for the α-
except the fully deuterated isotopologue). For experiments 4 and 5: all
isotopologues present (1−5).
Measurement of the Isotope Effect on Ethane Formation.
The assays contained 100 mM CH CH -S-CoM, 100 mM CD CD -S-
secondary IE (1.19 for Me-S-CoM and 1.22 for Et-S-CoM, per
3
2
D) points to an sp → sp hybridization change on going to the
transition state, but the β-secondary IE near unity excludes a
carbocation-like species.
3
2
3
2
CoM, and 4 mM coenzyme B in H O buffer. After addition of 42 nmol
2
Qualitatively, we judge the observed IEs to be compatible
enzyme (26.5 μM) per assay and closing the vial with the rubber
stopper, the headspace was evacuated inside the anaerobic tent. The
samples were incubated with the headspace under vacuum and
stopped by addition of 1.0 mL of air. The headspace gas was removed
with the radical substitution mechanism proposed by Siegbahn
1
7
and co-workers, but a more quantitative comparison must
await calculations of vibrational frequencies and IEs for both
substrates. However, the fact that allyl-coenzyme M does not
react or exchange isotopes at all, and that the relative energies
of TS1 and TS2 are different for the two substrates are difficult
to reconcile with this mechanism.
with a gastight syringe and passed through CDCl to analyze the
3
formed ethane for its isotopic constitution. For details of the
calculation of KIE on kprod see S4.3 in the SI.
Probing Propyl- and Allyl-Coenzyme M for Isotope
Exchange. Allyl-coenzyme M (10 mM) was tested for deuterium
While the primary KIEs for C−H activation (k /k = 2.4−
.5) are similar to that observed by Bergmann for the cleavage
in a putative alkane σ-complex, no additional information
H
D
incorporation in a D O assay according to the standard assay
2
2
conditions described above (enzyme: 15 nmol; 9.4 μM), incubation
time: 32 min at 60 °C).
22
1
4993
dx.doi.org/10.1021/ja4064876 | J. Am. Chem. Soc. 2013, 135, 14985−14995

Journal of the American Chemical Society
Article
Propyl-coenzyme M (10 mM) was tested using an assay containing
The simulated spectrum (sum of the lines calculated for all spin
systems) was overlaid with the acquired spectra, and each parameter
described above was fine-adjusted for each isotopologue.
Next, the population of each isotopologue was adjusted to match
the measured spectrum (see S2.4 in the SI for overlays of all fitted and
acquired spectra). The populations were normalized to mol % for
comparison with the COPASI simulation and for plotting.
Tables of all normalized values of isotopologues can be found in
S2.1 in the SI.
a Ti(III)-citrate in D O (see preceding publication for details
2
(
[Scheller, S.; Goenrich, M.; Thauer, R. K.; Jaun, B. J. Am. Chem.
Soc. 2013, 135, 10.1021/ja406485z]). Enzyme: 1.28 nmol (0.8 μM),
incubation time: 4 h at 60 °C).
1
H NMR analysis showed no deuterium incorporation for either
inhibitor.
Kinetic Simulation with the Program COPASI. For the ratio of
forward and backward reactions for substrate binding, the known
apparent K values are taken into account. The rate constant for ethyl-
Synthesis of Substrates. Labeled and unlabeled substrates were
synthesized as described previously.
M
8
−1
6,23
coenzyme M binding was simulated fast (10 s ), and the catalytic
step (step 3) was kept rate limiting. See SI (S3.1) for a detailed
description of the entire model.
Purification of MCR. Protein purification was carried out
following the procedure described in the preceding publication
CH CD -S-CoM was prepared from CH CD I (Cambridge Isotope
3 2 3 2
Laboratories). Scale: 1.72 g, 10.89 mmol; yield: 1.368 g, 7.04 mmol,
70%.
CD CD -S-CoM was prepared from CD CD I (Cambridge Isotope
3 2 3 2
(
Scheller, S.; Goenrich, M.; Thauer, R. K.; Jaun, B. J. Am. Chem.
Laboratories). Scale: 1.90 g, 11.5 mmol; yield: 1.107 g, 5.6 mmol, 56%.
Propyl-S-CoM was prepared from propyl iodide (Aldrich). Scale:
4.42 g, 26.0 mmol; yield: 3.715 g, 18.0 mmol, 72%.
Allyl-S-CoM was prepared from allyl iodide (freshly distilled). Scale:
4.39 g, 26 mmol; yield: 3.03 g, 14.8 mmol, 59%.
CD CH -S-CoM was prepared from CD CH I (Cambridge Isotope
3 2 3 2
Laboratories). Scale: 1.84 g, 10.6 mmol; yield: 1.077 g, 5.52 mmol,
55%.
Soc. 2013, 135, 10.1021/ja406485z). The concentration of the enzyme
solution (∼1 mL per purification) was ∼0.7−1 mM, corresponding to
2
0−28% by weight. For each preparation, it was determined exactly by
UV−vis spectroscopy. A typical UV−vis spectrum of an enzyme
preparation is provided in the Supporting Information of the
preceding paper (Scheller, S.; Goenrich, M.; Thauer, R. K.; Jaun, B.
J. Am. Chem. Soc. 2013, 135, 10.1021/ja406485z).
NMR Measurements. 1H NMR spectra were recorded on a
Bruker Avance II 600 MHz spectrometer with an inverse detection
ASSOCIATED CONTENT
Supporting Information
2
13
■
probe. H NMR spectra and C NMR spectra were recorded on a
Bruker Avance III 600 MHz spectrometer with a cryodetection probe.
The assay solutions were centrifuged and 620 μL transferred into an
NMR tube. After addition of 80 μL dioxane solution (0.025% v/v
dioxane and 0.025% v/v dioxane-d in D O) as an internal standard,
*
S
Additional figures, details of experimental procedures, and
acquired and measured NMR spectra. This material is available
free of charge via the Internet at http://pubs.acs.org.
8
2
the spectra were recorded at 278 K without spinning (30° excitation
pulse, 5 s acquisition time).
AUTHOR INFORMATION
Corresponding Author
jaun@org.chem.ethz.ch
■
Ethane was passed through CDCl with a fine needle and measured
3
with a 5 s relaxation delay (10 s repetition rate).
CW presaturation of the water signal for 2 s was employed for the
assays in H O and 2 s relaxation delay for the samples in D O to
2
2
Present Address
ensure a uniform repetition rate of 7 s for all samples.
§
Division of Geological and Planetary Sciences, California
Assays with deuterium-containing isotopologues were acquired with
1
2
Institute of Technology, 1200 E. California Blvd., Pasadena, CA
91125, United States.
and without broadband deuterium decoupling, denoted H{ H}.
2
1
H NMR measurements were carried out analogously to the H
2
NMR measurements, including D O presaturation where required. H
NMR measurements in undeuterated solvents (99% acetonitrile/1%
H O) were carried out without lock.
Notes
2
The authors declare no competing financial interests.
2
For the 13C NMR measurements with broadband 1H and 2
H
ACKNOWLEDGMENTS
■
decoupling (inverse gated), a relaxation delay of 60 s was employed to
allow reliable integration of the spectra.
This work was supported by the Swiss National Science
Foundation (S.S. and B.J.; Grant 200020-134476) and the Max
Planck Society and the Fonds der Chemischen Industrie (M.G.
and R.K.T.). We thank Marc-Olivier Ebert for his help with
special NMR techniques.
The integration was carried out with the software BRUKER
Topspin 2.1 relative to the internal standard dioxane (set to δ = 3.70
ppm for H and for H measurements, and set to δ = 67.0 ppm for
NMR measurements).
The exact isotopic constitution in the ethyl group of ethyl-
coenzyme M was determined by fitting of the spectra. The fraction of
conversion was determined by comparing the amount of disulfides
formed with the amount of remaining substrate, using the fittings of
the CH group at position 2: R-(S)-S-CH CH SO (for Et-S-CoM,
CoB-S-S-CoM, and CoM-S-S-CoM). The ratio of different ethane
1
2
13
C
REFERENCES
■
(1) Jaun, B.; Thauer, R. K. Met. Ions Life Sci. 2007, 2, 323−356.
(2) Hallam, S. J.; Putnam, N.; Preston, C. M.; Detter, J. C.; Rokhsar,
D.; Richardson, P. M.; DeLong, E. F. Science 2004, 305, 1457−1462.
(3) Hoehler, T. M.; Alperin, M. J.; Albert, D. B.; Martens, C. S.
Global Biogeochem. Cycles 1994, 8, 451−463.
(4) Gunsalus, R. P.; Romesser, J. A.; Wolfe, R. S. Biochemistry 1978,
17, 2374−2377.
−
2
2
2
3
isotopologues formed was determined by integration of the
corresponding H{ H} NMR spectra.
1
2
Fitting of NMR Spectra. Fitting of NMR spectra was carried out
with the software iNMR 4.1.7.
For each isotopologue, a spin system was defined for the ethyl
group. The chemical shifts were adjusted for each α and β deuterium
(5) Goenrich, M.; Mahlert, F.; Duin, E. C.; Bauer, C.; Jaun, B.;
Thauer, R. K. J. Biol. Inorg. Chem. 2004, 9, 691−705.
(6) Scheller, S.; Goenrich, M.; Mayr, S.; Thauer, R. K.; Jaun, B.
Angew. Chem., Int. Ed. 2010, 49, 8289−8292.
present (δ(CH ) = 1.1743 ppm, α-shift = −17.0 ppb, β-shift = −8.13
3
ppb; δ(CH ) = 2.5541 ppm, α-shift = −17.8 ppb, β-shift = −6.87
2
3
13
ppb). The proton coupling constant was J
= 7.4 Hz. The C-
(7) Ahn, Y.; Krzycki, J. A.; Floss, H. G. J. Am. Chem. Soc. 1991, 113,
4700−4701.
H,H
1
2
coupling constants for the CH group were: J = 127.3 Hz, J
4
addition, a C-α-shift had to be applied and the line width was
adjusted to the acquired spectra. (see S2.3 in the SI for a table of the
NMR data of each isotopologue)
=
3
C,H
C,H
1
2
.68 Hz; for the CH group: J
= 139.5 Hz, J =3.62 Hz. In
(8) Hoops, S.; Sahle, S.; Gauges, R.; Lee, C.; Pahle, J.; Simus, N.;
Singhal, M.; Xu, L.; Mendes, P.; Kummer, U. Bioinformatics 2006, 22,
3067−3074.
2
C,H
C,H
13
(9) Pine, M. J.; Barker, H. A. J. Bacteriol. 1956, 71, 644−648.
1
4994
dx.doi.org/10.1021/ja4064876 | J. Am. Chem. Soc. 2013, 135, 14985−14995

Journal of the American Chemical Society
Article
(
1
(
10) Matsson, O.; Westaway, K. C. Adv. Phys. Org. Chem. 1998, 31,
43−247.
11) Kogut, E.; Zeller, A.; Warren, T. H.; Strassner, T. J. Am. Chem.
Soc. 2004, 126, 11984−11994.
(
(
12) Geftakis, S.; Ball, G. E. J. Am. Chem. Soc. 1998, 120, 9953−9954.
13) Bengali, A. A.; Schultz, R. H.; Moore, C. B.; Bergman, R. G. J.
Am. Chem. Soc. 1994, 116, 9585−9589.
(
14) Schultz, R. H.; Bengali, A. A.; Tauber, M. J.; Weiller, B. H.;
Wasserman, E. P.; Kyle, K. R.; Moore, C. B.; Bergman, R. G. J. Am.
Chem. Soc. 1994, 116, 7369−7377.
(
(
15) Parkin, G. Acc. Chem. Res. 2009, 42, 315−325.
16) Parkin, G. J. Labelled Compd. Radiopharm. 2007, 50, 1088−
1114.
(
17) Chen, S.-L.; Blomberg, M. R. A.; Siegbahn, P. E. M. Chem.
Eur. J. 2012, 18, 6309−6315.
18) Scheller, S., Ph.D., Methyl-Coenzyme M Reductase: Mecha-
(
2
13
nistic Studies with H and C Labels. ETH Zurich, ETH Dissertation
19620, 2011, http://e-collection.library.ethz.ch/eserv/eth:2974/eth-
2974-02.pdf.
(
19) Ermler, U.; Grabarse, W.; Shima, S.; Goubeaud, M.; Thauer, R.
K. Science 1997, 278, 1457−1462.
(
(
20) Ahn, Y. Ph. D. Thesis, University of Washington, 1991.
21) Li, X. H.; Telser, J.; Kunz, R. C.; Hoffman, B. M.; Gerfen, G.;
Ragsdale, S. W. Biochemistry 2010, 49, 6866−6876.
22) Periana, R. A.; Bergman, R. G. J. Am. Chem. Soc. 1986, 108,
332−7346.
23) Noll, K. M.; Donnelly, M. I.; Wolfe, R. S. J. Biol. Chem. 1987,
62, 513−515.
(
7
(
2
1
4995
dx.doi.org/10.1021/ja4064876 | J. Am. Chem. Soc. 2013, 135, 14985−14995